The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the plugin-scanner package from PyPI using pipx install plugin-scanner if it is not already present. This is a functional requirement for the skill's primary purpose and targets the vendor's own package distribution.
[COMMAND_EXECUTION]: The skill uses shell commands to check for the existence of the scanner (command -v plugin-scanner) and to perform scanning operations (plugin-scanner scan, plugin-scanner lint, plugin-scanner verify). These are standard operational commands for a CLI-based security tool.
[INDIRECT_PROMPT_INJECTION]: The skill's core function is to ingest and analyze untrusted data from third-party repositories, SKILL.md files, and MCP servers. This creates a surface where malicious instructions embedded in the target files could attempt to influence the agent's behavior during the analysis phase.
Ingestion points: Data enters the agent's context via the plugin-scanner scan PATH command targeting external directories or repositories.
Boundary markers: The skill explicitly instructs the agent to treat findings as evidence rather than a guarantee of safety and warns against executing any code from the target repository.
Capability inventory: The skill has the ability to execute shell commands and read local file structures.
Sanitization: The skill relies on the external plugin-scanner binary to safely parse and interpret the content; no explicit sanitization logic is contained within the skill instructions themselves.

plugin-scanner